Higher Admissions, page 1
Higher Admissions
AN INITIATIVE OF THE
UNIVERSITY OF MICHIGAN
WITH SUPPORT FROM THE
ANDREW W. MELLON FOUNDATION
Earl Lewis and Nancy Cantor, Series Editors
Other books in this series:
Gary Orfield,
The Walls around Opportunity: The Failure of Colorblind Policy for Higher Education
Eboo Patel,
Out of Many Faiths: Religious Diversity and the American Promise
Scott E. Page,
The Diversity Bonus: How Great Teams Pay Off in the Knowledge Economy
Earl Lewis and Nancy Cantor, editors,
Our Compelling Interests: The Value of Diversity for Democracy and a Prosperous Society
Higher Admissions
The Rise, Decline, and Return of Standardized Testing
Nicholas Lemann
PRINCETON UNIVERSITY PRESS
PRINCETON & OXFORD
Copyright © 2024 by Nicholas Lemann
Commentary copyright © 2024 by Patricia Gándara
Commentary copyright © 2024 by Marvin Krislov
Commentary copyright © 2024 by Prudence L. Carter
Princeton University Press is committed to the protection of copyright and the intellectual property our authors entrust to us. Copyright promotes the progress and integrity of knowledge. Thank you for supporting free speech and the global exchange of ideas by purchasing an authorized edition of this book. If you wish to reproduce or distribute any part of it in any form, please obtain permission.
Requests for permission to reproduce material from this work should be sent to permissions@press.princeton.edu
Published by Princeton University Press
41 William Street, Princeton, New Jersey 08540
99 Banbury Road, Oxford OX2 6JX
press.princeton.edu
All Rights Reserved
Library of Congress Control Number: 2024932880
ISBN 9780691246765
ISBN (e-book) 9780691246772
Version 1.0
British Library Cataloging-in-Publication Data is available
Editorial: Eric Crahan, Rebecca Binnie
Production editorial: Elizabeth Byrd
Jacket: Katie Osborne
Production: Danielle Amatucci
Publicity: Alyssa Sanford (US); Kathryn Stevens (UK)
Copyeditor: Ashley Moore
Jacket Credit: dule964 / Adobe Stock
Contents
Introduction 1
Earl Lewis and Nancy Cantor
1 The Birth of the American Meritocracy 12
2 Higher Education for All 27
3 Testing, Affirmative Action, and the Law 45
4 Admissions without Testing 64
5 Testing without Meritocracy 78
Commentary: Higher Admissions,
a California Perspective 104
Patricia Gándara
Commentary: The Future of the Meritocracy 118
Marvin Krislov
Commentary: The Multiple Lives and Ironies of the SAT 130
Prudence L. Carter
Notes 143
Index 153
Higher Admissions
Introduction
By Series Editors Earl Lewis and Nancy Cantor
TESTING DEFINES LIFE for scores of students in the United States. According to a 2015 study by the Council of the Great City Schools, students take on average 112 standardized tests between kindergarten and when they graduate high school.1 While the study focused on young people in large urban districts, the sense that testing occupied hours per year held true for students in all districts. Some of these tests are formative, assessing progress and comprehension of a subject during a period of instruction. Other tests are summative and provide feedback after the period of instruction has ended. As Nicholas Lemann notes in this volume, over the last three generations one summative test has been elevated to national importance, the SAT—although in some states and communities the ACT has emerged as a competitor. Once called the Scholastic Aptitude Test, and now simply the SAT, this test has played a fundamental role in sorting the nation’s talent for more than seventy-five years. In the process we came to link success on the test with merit. This book and its comments call that logic into question.
The SAT first saw prominent use as a tool to democratize entrance into the nation’s Ivy League institutions. James Bryant Conant, president of Harvard University, worried that young men (it would be a generation before young women were included) who had not attended eastern boarding schools had a disadvantage when it came to taking the period’s College Board admissions exams. As a remedy, in the 1930s Conant, in partnership with Henry Chauncey, an assistant dean at Harvard, advocated for the adoption of the SAT, an adapted IQ test that had been developed experimentally by the College Board in the previous decade. The pair envisioned an admissions test that would allow them to move beyond the traditional candidate pool. In 1947 Chauncey became the first president of the Educational Testing Service, which began administering the SAT widely.
As Lemann reminds us in this important volume, critics emerged early on who worried that the SAT would wind up reducing access to college for people from disadvantaged backgrounds. An early critique came in 1948 from University of Chicago faculty members W. Allison Davis and Robert Havighurst. Davis, perhaps the country’s leading Black scholar, and Havighurst believed any form of IQ test rewarded social inheritance rather than measuring potential ability. For them race, socioeconomic status, and other social markers underpredicted the abilities of students from disadvantaged backgrounds on such tests, making the tests suspect.
Conant and Chauncey’s vision and Davis and Havighurst’s worries have animated the debate over high-stakes testing ever since. Some studies have repeatedly shown that standardized tests do not predict later success and only partially correlate with first-semester grades in college.2 Even a combination of test scores and grade point averages (GPA) functions as an imperfect predictor of anything more than first-semester grades.3 For a wide swath of the college-going population, they do not predict if a student will finish their course of study, graduate, and become a leader in their field or society more generally. Yet, as we witnessed in the recent Supreme Court cases on the use of race as one variable in the admissions process at selective colleges and universities, performance on one of the sanctioned college admissions tests (SAT or ACT, which stands for American College Testing, the SAT’s rival) factors mightily in how many think about merit.
Recent studies by Raj Chetty and colleagues confirm what many have long argued: performance on such tests is heavily influenced by socioeconomic status. Students from higher socioeconomic groups, irrespective of race, tend to fare better on the exams than students from lower socioeconomic groups.4 Several factors contribute to this pattern. Typically, more affluent families send their children to schools with greater educational resources—be they public or private. In addition, such families can afford the extra coaching high-achieving students often receive in advance of the testing. Finally, success on the test is openly valued in the family, in the surrounding community, and among one’s peers and friends.
Nor is race, regardless of class, an insignificant consideration when we chronicle the history of a test that became the test. Experimental social psychologist Claude Steele and students have produced and reproduced several studies that show you can lower performance on standardized tests by inserting key prompts before the test.5 Tell a Black or female student that they are representing not only themselves but their race or gender, and scores drop for individuals in each group who either are told or believe they are shouldering an individual or collective burden or responsibility. Steele sketches the effects quite beautifully in Whistling Vivaldi.
Trying to separate the psychocultural practices of test taking from effects on the tests means that an overreliance on one score, on one day, may result in the unwanted exclusion of talent. As Lemann notes, colleges and universities have sought to adjust for the possibility that they have unfairly diminished the pool by asserting that they seek to capture the total student. This has meant they scrutinize letters of recommendation, extracurricular activities, personal essays, and more.
In another book in the Our Compelling Interests series, The Walls around Opportunity, educational scholar Gary Orfield examines the interlacing of housing patterns, state and federal policy, demographic trajectories, and the failure of a colorblind approach in a racially stratified world. These factors produce obstacles to success, which Orfield labels the walls around opportunity in the United States. One of these walls, he maintains, is the wall around admissions to select colleges and universities, which returns us to the role and place of tests and testing.6 Unquestionably, schools are differentially resourced across the nation. This means colleges and universities have long needed ways of determining whether a grade at one school was equivalent to a grade at another school. Since the advent of mass admissions testing in the 1940s, first the SAT and later the ACT have served as such equalizers, purporting to standardize what we can determine about what students know and can demonstrate.
But the questions of sorting and testing have assumed even more political ballast in a world defined by ever-increasing racial and ethnic diversity. As William Frey noted in the series’ inaugural volume, we are on the verge of a diversity explosion.7 He and other demographers predict a nonwhite majority in the United States by 2040. This demographic trans
Notwithstanding the hundred or so tests school-age children take before graduating high school today, mass testing of children was not a mainstay when the SAT was created. After the Second World War, President Harry S. Truman formed a president’s commission to chart a future for American higher education. After deliberating for over a year, the commission issued its report in 1947.9 The Truman Commission recommended the democratization of access to college. It called for a massive financial aid program at the undergraduate and graduate levels, free tuition for those attending two-year schools, and a program of continuing education, and it conveyed the sense that education was neither purely utilitarian nor vocational. Reflecting a postwar ethos still being formulated, the commission, over some opposition, called for the end of segregation in the South and the ubiquitous use of quotas against Jews. Coming out of the war, the sentiment emerged that education would be the key to future progress and development and that access to such education needed to be widely available.
In time a tension emerged, one that moved standardized testing from the margins to the center. While the Truman Commission envisioned a democratized educational system open to all, others came to value and champion a select number of private and public institutions that would nurture a kind of talented elite. Most open-access institutions would not need a selection test. Anyone who applied got in. In states with growing populations, such as California, old methods no longer worked after the war. Certifying public high schools had long been a practice before the war in states like Michigan, where the University of Michigan had sent its faculty to certify school curricula as early as 1870. Students who passed a designated, certified curriculum gained admission. California came to adopt an education master plan, crafted by Clark Kerr, president of the University of California system, that directed students through one of three postsecondary doors—community college, a California State university campus, or a campus of the growing University of California system.10 In the last instance, a validated standardized test became a part of the admissions process.
Testing to control access grew in parallel with the divisions in postwar America. Tests that initially were used to expand the range of students admitted came to be associated with narrowing the range of students admitted. At one level it became a case of supply and demand: there were more applicants to some schools than available seats in the freshman class. At another level it is a story about the testing industry and the underlying paradigm that has prevailed for three-quarters of a century. Starting with Conant and Chauncey, the emphasis was testing for access rather than testing for success beyond first-year academic performance in college. If English is a second or third language for the test taker, what is the conventional admissions test measuring? What if we introduced a series of formative tests for applicants in their first year and allow those who master a subject area to go on?
A useful example may be drawn from the Meyerhoff Scholars Program, crafted by Freeman Hrabowski and the University of Maryland, Baltimore County. Hrabowski and colleagues learned that students in STEM (science, technology, engineering, and mathematics) fields graduated at higher levels if they got at least a B in introductory calculus and a couple of other core classes.11 Rather than selecting students for prior preparation, they modified the curriculum to allow students to repeat core introductory courses until they could show the needed mastery. This redesign amounted to a testing regimen for success and not just access.
The recent Supreme Court decision ending the use of race as a variable in the admissions process will undoubtedly drive new policies and innovations. As Lemann speculates, it could result in some schools abandoning admissions tests altogether. In recent years the test-optional movement has gained momentum across the United States, especially among elite liberal arts schools with national reputations. That movement raises the possibility that a class can be constructed without an overreliance on standardized tests. This prompts the question of how we have identified and defined merit and how we should henceforth.
Key to identifying the next steps to be taken is an understanding of the demographic reality pinpointed in the first volume in the Our Compelling Interests series and in Orfield’s volume The Walls around Opportunity. As Lemann notes, when the California Master Plan was introduced, the population across the state was 92 percent white; it is now 35 percent white. Orfield found that Black and Brown students account for the majority of public school students, yet the University of California system is nearly two-fifths white and one-third Asian.12 The passage of Proposition 209, which banned the use of race in the admissions process long before the recent Supreme Court decision, foretold what would happen to the makeup of elite campuses: except for a rise in the percentage of Asians, they would look more like they did in the 1960s—at least for a while.13 Some may believe that going backward is going forward but does doing so allow us to value and leverage diversity for the benefit of the nation and our democracy?
But as Patricia Gándara notes in her comment, the removal of test scores as a factor in the admissions process has been liberatory for Black and Brown students in California. Gándara was a member of a task force assigned to evaluate the continued use of SAT or ACT scores for admission into the University of California system of schools. Subsequently, the regents of the University of California recommended the abandonment of test scores and the use of thirteen other factors in the admissions process. The result was a surge of Black and Brown applicants and a noticeable increase in their presence on University of California campuses. She writes that after the decision to drop the SAT by the University of California Board of Regents, “Black freshman applicants rose by about 48 percent at both the University of California, Los Angeles, and the University of California, Berkeley, and Latino applicants increased by 33 percent at UCLA and 36 percent at Berkeley. This resulted in historically high rates of admission for these groups.” By the summer of 2023, while 40 percent of the state’s population, Latinos represented 22.5 percent of University of California students; Blacks, 6.5 percent of the overall state population, accounted for 5.5 percent of the University of California undergraduate population.
Looking ahead, we are left to ask, is diversity still a compelling state interest? Was the concern for generating a racially and ethnically diverse class at selective institutions a corollary to Conant and Chauncey’s desire to seed a meritorious elite? While rejecting one dimension of the diversity argument that has guided policy since the 1978 Bakke decision, the Supreme Court did not reject diversity as a compelling interest entirely. It carved out a special provision for the military academies to continue to identify and select recruits from all parts of the nation and to use race as a variable in the crafting of a class. Critics on the right complain that this is a zero-sum proposition. Race should never be a factor in admissions, even if there is a reason to believe it brings value.14
At some level the Court acknowledged what Scott Page argues in his book in our series The Diversity Bonus. According to Page, there are times when diversity matters less. If you need a lumberjack, he believes, you select the most capable, fittest lumberjack. But in a knowledge economy, where problems are often complex, research shows you gain a bonus when assigning a diverse set of actors to solving the problem. In such instances, an assembly of the smartest people from the same schools with similar backgrounds doesn’t produce the breakthroughs needed.15 The Court implicitly acknowledged that warfare and military strategy may benefit from pulling together a range of appropriately educated individuals and that the academies needed to generate that leadership. The decision seems to suggest diversity in the military would also contribute to group unity in the field.
